2 wireless access technologies...suring received signal level, delay profile characteristics, and...

1 Introduction

As clearly stated in the “e-Japan Project”,one of the most pressing issues for the nearfuture is the realization of a new-generation(NG) mobile communication system［1］ thatallows wireless access to IP networks (asexemplified by the so-called “ultra-fast Inter-net”) and, at the same time, permits seamlesshandover with existing wireless network sys-tems. Further, in the future, an enormous num-ber of access points (AP) will be available andcapable of establishing ultra-fast wireless con-nections to IP networks with carrier bit ratesexceeding 100 Mbps. These APs will accom-modate multiple mobile stations (MS) and willcontrol high-speed handover by procedurescoordinated using switches connecting the

APs, which will also execute the processesrequired for switching between heterogeneoussystems. The wireless transmission methodadopted for this purpose should adapt to anAP environment using software-defined wire-less technologies［1］to perform the appropriateselection of parameters concerning the wire-less transmission—such as the frequency,bandwidth, and transmission method.

One of the most essential technologies inthis NG mobile communication structure willbe a high-mobility, broadband, wide-areaaccess system that offers carrier bit ratesexceeding 100 Mbps under high-mobility con-ditions, with anticipated transmission speedsof several hundred Mbps to 1 Gbps in near-stationary environments, and that allows asmany users as possible to be accommodated

9HARADA Hiroshi and FUNADA Ryuhei

2 Wireless Access Technologies

2-1 Broadband Wireless Access System forNext Generation Seamless Mobile Communication

HARADA Hiroshi and FUNADA Ryuhei

In this article, we propose a new generation mobile communication system that can realize100 Mbps carrier bit rate under high mobility environment and can access IP network easily. Thenew system is based on dynamic parameter controlled orthogonal frequency and time divisionmultiple access (DPC-OF/TDMA) in which users share “slots” that use certain number of subcar-riers and certain time. To access the slots, mobile stations use a packet-reservation-based pro-tocol: packet reservation dynamic time-slotted multiple access (PR-DSMA). In addition, to avoidco-channel interference from adjacent cells and increase frequency utilization efficiency, we usean adaptive modulation scheme that is based on an interference detection algorithm. In this arti-cle, we mention the concept and the basic transmission performance of the proposed system.

KeywordsNew generation mobile communication system, OFDMA, TDMA, Beyond 3G system,1 cell reuse

10 Journal of the National Institute of Information and Communications Technology Vol.53 No.4 2006

on a single AP.We may identify a number of specific

requirements for such a high-mobility, broad-band, wide-area access system: (1) the cellradius for the frequency band used should beextended to the full extent possible, whichmeans increasing the number of users to thefull extent as well; (2) the efficiency of fre-quency use should be maximally enhanced, toavoid unnecessary occupation of frequencybands when expanding the communicationarea; (3) low packet-error rates should beensured in transmission, even under high-mobility conditions; (4) easy access must beoffered to IP networks (such as the Internet);(5) the required band must be supplied to theuser based on considerations such as userdevice environment, cost effectiveness, bandlimitations, etc. for all users—from thoseusing PDAs to those using vehicle-bornedevices; and (6) each of the layers (such as theseven layers of the OSI) must be distinguishedand categorized, and the lower layers must becontrollable from the upper layers.

In particular, a minimum carrier rate ofseveral tens MHz (even under high-mobilityconditions) and a maximum rate of 100 Mbpsare required under condition (1) above, and soa frequency band of approximately 100 MHzmust be secured. If we also take into consider-ation the expansion of the cell radius, then thefrequency band used will mainly be in themicrowave band of 3－10 GHz. However,availability in the microwave band is presentlyextremely low, and so the same frequencymay have to be used for adjacent cells whenexpanding the communication area with theadoption of cellular systems. In such a case,the cell boundaries are subject to interferencefrom multiple cells, and so countermeasuresmust be devised to counteract this interfer-ence. One method that has been examined forthis purpose is referred to as Code DivisionMultiplexing (CDM). Another method is seenin OFCDM transmission［2］, which offersadvantages over CDM in the suppression ofthe frequency used; this method also adoptscoded Orthogonal Frequency Division Multi-

plexing (or “coded OFDM”), which providesfor sufficiently high transmission quality evenunder high-mobility conditions.

However, previous studies［3］-［5］—mea-suring received signal level, delay profilecharacteristics, and building penetration char-acteristics at microwave frequencies (4.9-GHzband) planned for use in NG mobile commu-nication systems—were conducted under thefollowing conditions: a reception antenna gainof 4.7 dBi (omnidirectional in the horizontalplane), a transmission power of 40 dBm, atransmission antenna gain of 9.7 dBi (omnidi-rectional in the horizontal plane), and a trans-mission antenna height of 40 m. According tothe results, the radius of reception points atwhich the gain of the received signal level of -80 dBm can be obtained is 500 m at maxi-mum; the median value of the delay profile is300 ns, the 90 % value is 450 ns, and the max-imum value is between 3,000 and 4,000 ns.The building penetration values are approxi-mately 40－45 dB, and the areas in whichinterference arises between adjacent outdoor/outdoor cells and between outdoor/indoorcells are not large compared to that of secondand third-generation mobile communicationsystems. Thus, we may conclude that the samefrequency can be used in adjacent cells notonly by adopting a method combining CDMand OFDM, but also by simply combininginterference detection and circumventionmethods with OFDM transmission.

In this article, we propose a DynamicParameter Controlled Orthogonal Frequencyand Time Division Multiple Access (DPC-OF/TDMA) system, a high-performance sys-tem resulting in carrier bit rates exceeding100 Mbps even in high-mobility conditions,and offering superior connectability to IP net-works［6］［7］. The proposed system has OFDMtransmission as its core, and “slots” havingprescribed lengths of frequency and time areshared between multiple mobile stations (MS).The method used to access these slots uses aprotocol for packet reservation dynamic time-slotted multiple access (PR-DSMA) that isbased on reserved dynamic slot allocation


technology. Further, the system uses an inter-ference detection and adaptation modulationmethod for interference control, with the aimof increasing transmission volume and effi-ciency of frequency use. In this article, wewill focus on the basic concepts and featuresof the proposed system.

2 Outline of the system

The DPC-OF/TDMA has as its basic con-cept the adoption of the OFDM transmissionmethod, as shown in Fig. 1. In order to carryout OFDM transmission, the subcarrier isdivided into blocks having prescribed lengths.Each of these blocks will be referred to as sub-

Fig.1 Frame configuration of the DPC-OF/TDMA for (a) downlink and (b) uplink


channels in this article. Similarly, the time axiswill also be divided into blocks of prescribedlengths. The blocked “subchannels” will bereferred to as “slots”, and these slots will beshared among multiple MS. When an MSestablishes communication with APs, it regis-ters its own station to the subchannel, and theslot-allocation controller of the AP allocates atime slot within the subchannel. The slot- allo-cation information of each subchannel isinserted periodically in the form of a FrameControl Message Slot (FCMS) in the timeaxis, as shown in Fig. 1. The slots betweenone FCMS and the next are referred to belowcollectively as the “TDMA frame”.

In this method, an MS does not need to beable to access all subchannels, as shown inFig. 2, and the number of subchannelsaccessed may be specified based on the scale,power consumption, and available servicecharge options for the MS. Moreover, evenwhen it is possible to access all subchannels,the number of such actually used may be var-ied and controlled by parameters set accordingto conditions specified by the user—relating,

for example, to power consumption of the MSdevice and service charges. Further, the modu-lation method for each slot may also be variedthrough parameter settings. Nevertheless, itwill remain preferable that receiving ends onall MS be capable of receiving all subchannelsso that these reception points may efficientlysearch through these subchannels for emptyslots.

3 Configuration of the physicallayer

Figure 3 shows the configuration of thephysical layer of the receiver unit in the APand MS. Below, transmission from the AP tothe MS will be referred to as “downlink”, andthe reverse transmission will be referred to as“uplink”.

During downlink, the AP receives aTDMA frame for each subchannel from theMAC layer and allocates each of these frames,as is, into the available subchannels. In addi-tion to this allocation data signal, pilot signalsfor radio propagation path estimation and a

Fig.2 Conceptual drawing of use of proposed system


preamble signal containing information on themodulation method will be inserted. The basicpacket format will be as shown in Fig. 1.Next, each subcarrier will be subjected tomodulation such as BPSK and QPSK (map-ping) based on commands from the MAClayer, and the modulated signal for OFDMwill then be created through the execution ofIFFT. Guard intervals will then be inserted

and the frame will be up-converted (U/C) intoan analog signal and finally converted intosignals in the given transmission frequencyband, for transmission to each MS.

At the MS, the signal is down-converted(D/C) to the baseband and synchronized afterdigitization, and FFT is performed. The fre-quency divisionsignals are then divided andallocated among the subchannels. A radio

Fig.3 Configuration of the physical layer in the transmission and receiver unit [(a) AP and (b) MS]


propagation path estimation is then performedusing the pilot channels within the preamble,and a correction to the transmitted signal ismade based on the results of estimation. Sub-sequently, each subchannel signal is handedover to the MAC layer. The MAC layerreceives the slot information for each sub-channel based on the MAC protocol (whichwill be described in a later section) andreceives the required information of the slotfor each subchannel and sends the informationto the higher layers.

During uplink, on the other hand, TDMAtransmission is performed within the time pre-scribed by each MS based on the slot-alloca-tion method for each subchannel in downlink.At each MS, the MAC layer receives theTDMA frame for each subchannel, and allo-cates each of these frames as is to the avail-able subchannels. To this information the pilotsignal and the preamble signal containing themodulation information are added, and modu-lation is then performed on each subcarrier inaccordance with commands from the MAClayer. IFFT is then performed to create a mod-ulated signal for OFDM. In this process, it isassumed that the numbers of subchannels andthe IFFT are appropriate for the number ofsubchannels accessed by the MS, based onconditions such as scale, power consumption,and charge options applicable to the MS.However, it is possible to use an IFFT of thesame scale on the transmission side of the AP.

The AP receives signals from multipleMSs. After the downlink slot information hasbeen received, the uplink signals are transmit-ted within a prescribed length of time; howev-er, export time may differ among the signalsfor each MS. Thus, synchronization isrequired to perform FFT on multiple signalshaving these different exportation times. Thus,the question of the size of the FFT used mustbe resolved. One method we may considerinvolves synchronization according to the ear-liest arriving signal, or according to the largestsignal, and to perform batch FFT on each sub-channel using an FFT of a scale sufficient tohandle all subcarriers［8］. Another method is to

prepare an FFT for each subchannel based onthe allocated channel number, and then to per-form the FFT. As shown in Fig.1, the configu-ration for each slot is basically the same as forthe uplink case, except for certain differencesin the parameters.

The APs must be capable of high-qualityreception of information from high-mobilityMS, or for the MS to receive information fromthe AP even in high-mobility environments.Since the symbol for channel estimation underDPC-OF/TDMA is located at the head of thetransmitted packet, information located closeto these symbols features relatively low errorrates. However, with longer packets and high-er mobility, the error rate increases for infor-mation located farther from the symbol. Toresolve this problem, the DPC-OF/TDMAadopts a sequential channel estimationmethod［9］［10］. Under this method, signalsreproduced on the receiving end are used tocreate a replica of the received signal (afterFFT), as shown in Fig. 3. This replica is thencompared to the received signal after FFT toconfigure the channel characteristics estimatedbased on the packet header.

4 Architecture of the MAC layer

4.1 Basic architectureAs shown in the architecture of the MAC

layer in Fig. 1, a common protocol is used foreach subchannel. The protocol presentlyadopted is the PR-DSMA protocol, which isbased on reserved dynamic slot allocationtechnology［11］. This protocol utilizes theFDD (Frequency Division Duplex) method,which uses different frequency bands fordownlink and uplink. Further, the multipleaccess method is based on the TDMA method,in which the AP performs time slot allocationfor each mobile station according to the trafficconditions.

4.2 Frame formatAs shown in Fig. 1, the downlink frame

consists of a single FCMS (Frame ControlMessage Slot) and a number n of MDS (Mes-


sage Data Slot). The uplink frame consists of asingle ACTS (Activation Slot) and severalMDS (The number of ACTS plus the numberof MDS will equal n). Here, the frame fre-quency for downlink and uplink is the same.Each of the various slot types will bedescribed below in more detail.(1) FCMS (Frame Control Message Slot)

This slot is specialized for downlink, andeach frame will have one FCMS, which is nor-mally located at the head of the frame. Theslot elements are composed of the headercomponent—which consists of the slot type,mode, version, slot count, sub slot count,source MAC address, destination MACaddress, channel count, slot assignment term,propagation delay, transmit control, accesspoint identification, and header CRC—and thepayload component—which consists of theslot assignment information for downlink, slotassignment information for uplink, ACKinformation for uplink, and payload CRC［11］.(2) MDS (Message Data Slot)

One MDS slot (or more) is allocated to asingle downlink or uplink frame. In the down-link, the AP performs the multiplexing, whilein the in the uplink, multiple MSs do so. TheMDS is used in normal data transmission, aswell as for registration/deregistration requestsby AP and connection establishment/releaserequests in the downlink. In the uplink, thisslot is used for ACK transmission for down-link data. The slot elements are composed of aheader component—which consists of the slottype, packet type, control, buffer count, sourceMAC address, destination MAC address, datalength, data sequence number, ACK sequencenumber, and header CRC—and the payloadcomponent—which consists of the payloadand payload CRC［11］. In this method, thePDU (packet data unit) from the higher hierar-chy of the MAC layer is not sent directly bythe MDS; instead, the PDU is divided intosegments of a certain length, and the segmentsare then sent via MDS. (3) ACTS (Activation Slot)

This slot is mainly allocated to an uplinkframe, and serves as a randomly accessible

slot used for transmitting requests for registra-tion/deregistration by MS and connectionestablishment/release. As shown in Fig. 1, theslot in fact consists of several mini-slots; whenthe MS issues a request to the AP, a slot isselected randomly from these mini-slots andthe requested packet is sent within the timeselected. The slot elements are composed ofthe header component—which consists of theslot type, packet type, control, data length,source MAC address, destination MACaddress, and header CRC—and the payloadcomponent—which consists of the payloadand payload CRC［11］.

4.3 Basic communication protocol［11］Figure 4 is a block diagram of the MAC

layer of each subchannel. In the AP, all of theIP data to be transmitted are divided into seg-ments having a certain length according to theslot lengths, and stored in the transmissionbuffer of the AC according to the registeredMS. In the slot allocation unit, the MDS dataare created based on the specified slot alloca-tion method, and combined with the FCMSdata generated by the control data configura-tion unit to produce a TDMA frame. Theframe is then sent to the MS.

At the MS, the FCMS is extracted fromthe TDMA frame sent from the AP in the slotextractor, and the downlink/uplink allocationstatus is checked for each subchannel. Thetraffic conditions are then detected. When per-forming downlink/uplink, the channel selectorof the MS relays its intent to use the respectivesubchannel to the physical layer, and the sig-nal from the MS registration signal generatoris sent to the base station using ACTS. Oneexample of the MS registration signal may beseen in the MAC address of the MS.

When the AP is able to receive the ACTSinformation, the AP, which has received therequest to use a subchannel from the MS,issues the name of the authorized MS and thetemporary address to all of the MS using theMDS for downlink. The relevant MS uses theaddress to initiate a session with the AP.

In this method of communication, a single


MDS slot is allocated in a round-robin mannerto the MS that is registered to the AP, regard-less of the existence of the uplink transmissionbuffer. In this case, if the uplink transmissionbuffer status for the MS is 0 and no valid datahas been sent by the MS within a certainlength of time after the slot has been allocated,no slots will be allocated to for a given lengthof frame time. The MS will use this allocateduplink MDS slot to request and send informa-tion.

4.4 Slot allocation method4.4.1 Downlink

Figure 5 shows the slot allocation methodfor the downlink. The TDMA frame used inthe downlink contains slots for FCMS andMDS, and the FCMS will always be allocateda single slot at the head of the frame. TheMDS consists of the MDS for MS data trans-mission and MDS for call-control data trans-mission, to send information required for callcontrol. When call-control data needs to be

Fig.4 Architecture of the MAC layer transmission/reception unit [(a) AP and (b) MS]


sent, these data will be preferentially be allo-cated slots from the head of the frame, relativeto other MS data.

An MS that has entered the communica-tion zone of a certain AP will first carry outregistration with the AP using ACTS. Whenthe AP has received the registration request, itis relayed by the FCMS, and at the same time,the request for communication is broadcastusing the MDS. The slot allocation for the MSdata transmission MDS is then made, accord-ing to the higher-layer PDU (packet data unit)length for each PDU, following the order oftheir entry into the transmission buffer. Whenthere is no unsent data in the transmissionbuffer, no slots are allocated, and when onePDU is sent, an uplink MDS is allocated. ThisMDS is used to transmit ACK; thus constitut-ing transmission of downlink information.Figure 5 shows a case in which allocation ismade for downlink when a 1,500-byte PDU issent to two MS, A and C. In this case, it isassumed that the volume of data transmittedby each MDS is 128 bytes.4.4.2 Uplink

Figure 6 shows the slot allocation methodfor the uplink. The TDMA frame used in theuplink contains slots for ACTS and MDS, the

latter of which can be categorized into thatused for MS data transmission and that usedfor transmitting ACK for downlink data(ACK-data MDS). The slot allocation forACK data MDS is always performed when thepacket within the MDS sent in the downlinkcorresponds to the last segment in the higher-hierarchy PDU. The slot position assigned toACK-data MDS will be later than thatassigned to the slot for transmitting the lastsegment in the higher-hierarchy PDU in thedownlink.

The slot allocation for the user-data MDSwill use the slots remaining after the allocationto the ACK-data MDS. The slot-allocationmanagement table will be referred to, and theslots will be allocated in a round-robin mannerto the MSs that have established connections.The round-robin method here will not makeallocations per PDU, as is the case in down-link, but will instead make allocations per seg-ment. Figure 6 shows a case in which alloca-tion is made for uplink when a 1,500-bytePDU is sent to three MS—B, D, and E. In thiscase, it is assumed that the volume of datatransmitted by each MDS is 128 bytes.

Fig.5 Slot allocation method for downlink


4.5 Re-transmission control methodIn PR-DSMA, re-transmission control

using Go-back-N may be used, as shown inreference［11］. Note that re-transmissions arenot made for broadcast data for transmissionperformed by MDS in the downlink or forMDS data that contain only ACK/NACK datain the uplink.

5 Evaluation using computer simulations and prototype testing

To evaluate the transmission characteris-tics of the proposed communication system,we have carried out computer simulations andhave also constructed a prototype.

5.1 Example of frame configurationIn the proposed DPC-OF/TDMA, the

basic frame configuration for transmissions atcarrier rates exceeding 100 Mbps were inves-tigated using the parameters in Fig. 1. Table 1shows an example of the results. For the para-meters in this table, the temporal lengths ofMDS and FCMS are the same, and the lengthof the ACTS is half of that for MDS and

FCMS. When a single FCMS and eight MDSare sent in the downlink and four ACTS andseven MDS are sent in the uplink, the tempo-ral length of a single TDMA frame was thesame for uplink and downlink.

5.2 Example of basic transmissioncharacteristics of MDS

An evaluation was made of transmissionproperties, with special emphasis on slot errorrate, when performing MDS transmissionusing all subchannels using the parameters inTable 1. The frequency band was 5 GHz, andthe Doppler frequency was assumed to be500 Hz (corresponding to 100 km/h). Thechannel model used was a 24-wave exponen-tially attenuating independent Rayleigh fadingmodel (delay spread of 350 ns). The basicscheme of the MDS frame is shown in Fig. 7,and the results are presented in Fig. 8. Withthese parameters, it was assumed that theOFDM signals (having a total carrier numberof 768 and subcarrier number of 64) dividedinto 12 subchannels were all used for thetransmission. Two OFDM symbols for radiopropagation characteristics estimation wereprepared in addition to those for MAC data

Fig.6 Slot allocation method for uplink


transmission, and the symbols were used toacquire radio propagation characteristics andto alleviate the effects of distortion of thetransmission signals caused by fading. TheQPSK modulation method was employed. TheMDS error rate shown in Fig. 8 may bedefined as the ratio of erroneous MDS (i.e.,those that contain even a single error in data)when information is transmitted via MDS asshown in Fig. 7. From this figure, it can beseen that the MDS error rate obtained for thecase when QPSK modulation is used as theprimary modulation method at a speed of100 km/h for the proposed method is on the

order of 10e-2 at Eb/N0 greater than 10 dB. Inthe case of QPSK, the volume that can betransmitted by one MDS is 128 kbytes. There-fore, an IP packet having maximum length canbe sent using 12 MDS. In other words, whenthe MDS error rate is on the order of 10e-2,then the error rate for a maximum-length IPpacket is at most 10e-1, and so informationtransmission is thus possible through re-trans-mission of packets.

5.3 Evaluation of the prototypeTo investigate the feasibility of the pro-

posed DPC-OF/TDMA hardware, we havedeveloped the prototype shown in Fig. 9 usingthe parameters in Table 1. The specificationsof the prototype are presented in Table 2. Thefrequency used was 4 GHz on the base stationside, and 3 GHz on the mobile terminal side.The transmission powers were 5.5 W and 5 Wfor the base station and mobile terminals,respectively. The present prototype used a

Table 1

Fig.7 General scheme of the MDS frame

Fig.8 MDS error rate


768-carrier OFDM and the 99 % occupiedbandwidth was approximately 96 MHz, asshown in Fig. 10. The transmission character-istics of this prototype were measured for fad-ing conditions similar to those used in thecomputer simulation in Section 5.2 , byincreasing the speed generated by the hard-ware fading simulator. QPSK was used as theprimary modulation method, and when a codehaving a coding length of 1/2 was used (corre-sponding to a maximum carrier bit rate of76.8 Mbps), the IP data throughput was34.026 Mbps for downlink when the transmis-

sion and receiver units were directly connect-ed, and the maximum throughput was 21.467Mbps when 24-wave exponential-attenuatingmulti-path fading was applied. For the uplink,the IP data throughput was 30.02 Mbps whenthe transmission and receiver units weredirectly connected, and the maximumthroughput was 16.59 Mbps when 24-waveexponential-attenuating multi-path fading wasapplied. Thus, it was proven that a maximumIP data throughput of at least 10 Mbps couldbe secured even at high mobility of 100 km/h,for both downlink and uplink.

Fig.9 General scheme of prototype

Fig.10 Occupied bandwidth


6 Conclusions

This article proposed a new-generationmobile communication system DPC-OF/TDMA based on OFDM and providedsummaries of the system’s physical layer andthe MAC layer. We evaluated the transmissioncharacteristics of the system by computer sim-ulations and prototype testing, thus provingthat the present system is capable of providinga minimum carrier bit rate of several tens ofMbps even in a multi-path environment withhigh mobility exceeding 100 km/h. This eval-uation also demonstrated the possibility ofoffering a carrier bit rate of up to several hun-

dred Mbps. The present prototype was the firstin the world to succeed as a mobile communi-cation system employing OFDMA and TDMAat carrier bit rates greater than 100 MHz, hasproven itself capable of becoming the fourth-generation mobile communication system forthe future, and will no doubt form one of thetouchstones of future research. In the future,our group hopes to pursue the standardizationof various technologies that have been gainedin the course of obtaining these results, and toconduct further and more detailed transmis-sion tests using the system in outdoor environ-ments.

Table 2 Prototype specifications

References01 H. Harada, M. Kuroda, H. Morikawa, H. Wakana, F. Adachi, “The Overview of the New Generation

Mobile Communication Systems and the Role of Software Defined Radio Technology”, IEICE Trans.

Commun., Vol.E86-B, No.12, pp.3374-3384, Dec. 2003.

02 H. Atarashi, S. Abeta, M. Sawahashi, “Variable Spreading Factor-Orthgonal Frequency and Code Divi-

sion Multiplexing (VSF-OFCDM) for Broadband Packet Wireless Access”, IEICE Trans. Commun.,

Vol.E86-B, No.1, pp.291-299, Jan. 2003.

03 H. Iwai, K. Yonezawa, T. Maeyama, H. Shinonaga, H. Harada, M. Fujise, “Path Loss Measurement in

5 GHz Macro-Cellular Environments”, IEICE Conference vol.B-1-46, pp.46, Mar. 2003.

04 K. Yonezawa, T. Maeyama, H. Iwai, H. Harada “Delay spread measurement in a 5GHzmicro cellular sys-

tem”, IEICE Communication Society Conference, B-1-39, Sep. 2003.

05 K. Yonezawa, T. Maeyama, H. Iwai, H. Shinonaga, H. Harada, M. Fujise, “An Experimental Study on

Penetration Characteristics from Outdoor to Indoor at 5 GHz Band", IEICE Conference 2003 vol.B-1-44,

pp.44, Mar. 2003.


06 H. Harada, C-J Ahn, S. Takahashi, Y. Kamio, S. Sampei, “New Generation Mobile Communication Sys-

tem by Dynamic Parameter Controlled OF/TDMA”, Technical report of IEICE, RCS2003-284, pp.41-46,

Jan. 2004.

07 H. Harada, C-J Ahn, S. Takahashi, Y. Kamio, and S. Sampei, “Dynamic Parameter Controlled Orthogo-

nal Frequency and Time Division Multiple Access”, PIMRC 2004, vol.4, pp.2648-2652, Sep. 2004.

08 R. Kimura, R. Funada, H. Harada, S. Shinoda, “A Study on Complexity Reduction Scheme for Time

Alignment in Uplink Dynamic Parameter Controlled OF/TDMA”, Technical report of IEICE, RCS2004-439,

pp.43-48, Nov. 2004.

09 H. Harada, R. Funada, C-J Ahn, S. Takahashi, M. Sawada, “OFDM Packet Transmission System with a

Sequentially Updated Channel Estimation Algorithm : Concept and Fundamental Transmission Charac-

teristics”, Technical Report of IEICE, RCS, pp., Jan. 2005.

10 R. Funada H. Harada, Y. Kamio, S. Shinoda, M. Fujise, “A New Amplitude and Phase Compensation

Scheme under Fast Fading Environment for OFDM Packet Transmission Systems”, VTC 2001 Fall,

pp.2093-2097, Oct. 2001.

11 H. Harada, “100 Mbps and beyond point-to-multipoint ultra high-speed radio access system : concept

and fundamental experimental results”, VTC2003-Fall, Vol.3, pp.1874-1879, Oct. 2003.

HARADA Hiroshi, Ph.D.

Research Manager, Ubiquitous MobileCommunication Group, New Genera-tion Wireless Communications ResearchCenter

Cognitive Radio, Software DefinedRadio, Broadband Wireless Access Sys-tem

FUNADA Ryuhei, Ph.D.

Expert Researcher, Ubiquitous MobileCommunication Group, New Genera-tion Wireless Communications ResearchCenter

Cognitive Radio, Software DefinedRadio, Broadband Wireless Access Sys-tem

2 wireless access technologies...suring received signal level, delay profile characteristics, and...

Documents