| 520 |
|
|
|a SAŽETAK: Prilagodljivo skakanje frekvencija u OFDM širokopojasnim radijskim mrežama Kontinuirano povećanje količine multimedijskih sadržaja na Internetu i pojava novih usluga zahtijevaju sve veće brzine prijenosa podataka i mogućnost pristupa s bilo koje lokacije. Pristup neovisan o lokaciji, bilo mobilni ili nomadski, omogućavaju radijske mreže. S obzirom na specifične uvjete koji vladaju u zraku kao prijenosnom mediju, tehnološke inovacije idu u smjeru povećanja brzine prijenosa podataka uz štedljivo korištenje raspoloživog prijenosnog pojasa. Kako bi se postigli navedeni ciljevi razvijen je model višekorisničkog sustava baziran na OFDM (Orthogonal Frequency Division Multiplex) postupku i WiMAX (Worldwide Interoperability for Microwave Access) normi uz dvije značajne nadogradnje. Prvu predstavlja algoritam za raspodjelu resursa korisnicima na razini podnosioca, a drugu rješenje bazirano na dinamičkom određivanju primijenjene modulacije i kodiranja te skakanju frekvencija. Primijenjeni algoritam primarno radi na principu fer raspodjele resursa uz dodatnu komponentu maksimizacije brzine. Prema informacijama o stanju prijenosnog kanala, dinamički se odabiru najbolji raspoloživi podnosioci za svakog korisnika posebno. Pri raspodjeli uvažavaju se QoS (Quality of Service) zahtjevi vezani za vrstu prometa, daje se prednost prometu u stvarnom vremenu, a zatim se resursi dodjeljuje ostalom prometu. Kako bi se provjerile značajke predloženog algoritma izvršena je usporedba s postojećim WF (Water Filling) i PF (Proportional Fairness) algoritmima. Promatrana je sposobnost zadržavanja jednolike raspodjele resursa između korisnika te ukupni kapacitet sustava pri značajnoj razlici u kvaliteti prijenosnog kanala. Ukupni kapacitet sustava pri niskim SNR (Signal to Noise Ratio) vrijednostima i velikim razlikama u kvaliteti prijenosnog kanala nešto je manji nego kod WF i PF algoritama, ali je zadržana jednolika raspodjela između korisnika. Dodatne razlike u odnosu na postojeće sustave su određivanje modulacije i kodiranja na razini svakog podnosioca, dinamičko određivanje predloška za skakanje frekvencija i izravno dodavanje podnosilaca korisnicima. Ovako koncipiran sustav uspoređen je s postojećim FUSC (Full Usage of Subchannels) i AMC (Adaptive Modulation and Coding) permutacijskim shemama. Dobiveni rezultati pokazuju dominaciju predložene metode u odnosu na FUSC i AMC u slučaju loših uvjeta na prijenosnom kanalu. Kombinacija prilagodljive modulacije i kodiranja uz skakanja frekvencija prema dinamički određenom predlošku predstavlja moguće rješenje za postizanje većih brzina prijenosa informacija i stabilnijeg rada u slučaju loših uvjeta na prijenosnom kanalu. -
|b KLJUČNE RIJEČI: prilagodljiva modulacija, dinamičko dodjeljivanje resursa, skakanje frekvencija, OFDMA, QoS, raspoređivanje podnosilaca
|
| 520 |
|
|
|a ABSTRACT: Wireless networks enable nomadic or mobile access to Internet. Data throughput and transmission quality in wireless networks are, however, significantly lower than in fixed lines due to lack of bandwidth and complex conditions in the transmission media – air. The continuous increase of multimedia content enhances the needs for higher data speed in communication channels. Many advanced technologies such as orthogonal frequency division multiple access (OFDMA), multiple-input and multiple-output (MIMO), frequency hopping (FH), adaptive modulation or dynamic resource allocation, are used to solve this problem. A model of transmission communication system based on OFDM and the 802.16 standard, upgraded with resource allocation algorithm (RAA) and FH is described and analysed in the doctoral thesis. The thesis is split into five chapters while supplements to the thesis (A, B, and C) contain the major part of the simulation code written in MATLAB Simulink. After the first, introductory, part, the second chapter deals with a historic overview of mobile networks from first generation analogue networks (1G) to present broadband mobile networks (4G). A special attention is drawn to broadband networks defined by the 802.16 standard. Their rapid development started in 2001, with support for fixed broadband networks first, that was expanded for mobile users in 2005. Present 802.16m version is one of possible solutions for 4G networks. Despite the differences in these two standards, both of them use the same or similar technologies based on OFDM. Theoretic description of OFDM and WiMAX standard, FH and RAA are elaborated. Chapter 3 outlines a developed system model based on the 802.16 standard. The OFDMA transmission system is used with 192 data subcarriers, 8 pilots, 256-point FFTs, cyclic prefix length ⅛ and channel bandwidth of 3.5 MHz. Adaptive modulation and coding are applied according to the conditions in the transmission channel. Seven MC groups are used for modulation and coding in the simulation: OFDMA with BPSK ½, QPSK ½, QPSK ¾, 16QAM ½, 16QAM ¾, 64QAM ⅔ and 64QAM ¾. The coding is run as Forward Error Correction (FEC), consisting of a RS outer code concatenated with a rate-compatible inner CC. The data are subsequently coded, modulated, packaged and prepared for sending. After that the data pass through three independent channels until they are received by three receivers, decoded and demodulated. Stanford University Interim (SUI), SUI 3 channel model is used in the simulation. A non-fading, flat-fading, or dispersive multipath fading can be chosen in each of three independent transmission channels. Additive white Gaussian noise (AWGN) variance (in SNR mode) is added to the signal. An appropriate K factor, maximum Doppler shift, number of paths, and path gains can be changed accordingly. The data rate varies dynamically in relation to the channel conditions. As subcarriers are allotted to each user independently, it was necessary to develop an adequate RAA. The proposed RAA is a two-phase algorithm. The incoming data contain real time (RT) and non-real time (NRT) data traffic. In the first phase the algorithm allocates RT and, if defined, mandatory NRT (mNRT) data traffic. The resources for mNRT data are assigned to ensure that all users get minimum guaranteed bandwidth for NRT data traffic. In the second phase NRT data traffic is distributed to its subcarriers up to the full system capacity. Due to stringent QoS requirements for RT traffic, the subcarriers with the best SNR values are respectively selected for each user. The assignment of better quality subcarriers to RT traffic guarantees the satisfaction of QoS requirements. The entire system capacity is enhanced in this way and the required QoS parameters (e.g. small delay, delay variation, and maximum sustained data rate) are achieved. The weight factors are direct measures of assigned throughput of a user, expressed in allocated bits’ number for useful data transmission in the observed time period within a particular frame. Fair resource allocation is achieved by equalizing weight factors, while data rate maximization is accomplished by selecting the best subcarriers for each user. The algorithm for dynamic FH is embedded in the system. FH is done if there is a free subcarrier of the same quality for an observed user. The system selects the subcarrier that was not used in the previous frame. Frequency diversity is enhanced in this way. As FH is done exclusively among the same quality subcarriers, there is no reduction in throughput and transmission rate of the system. Hopping pattern for FH is not pre-defined because it is determined dynamically based on the subcarriers’ quality. The results of the simulation are analysed in chapter 4. The features of the proposed algorithm are compared in three segments: resource allocation, achieved system capacity and BER. Resource allocation between the users in the proposed algorithm is uniform in broader area of SNR values compared to water filling (WF) and proportional fairness (PF) algorithms. Total system transmission capacity is compared for three users under different channel quality conditions by applying the proposed, WF and PF method. The WF method continuously ensures maximum system capacity, but with disproportional resource allocation between the users. The PF algorithm provides higher system transmission capacity compared to the proposed method at lower SNR values, but with worse, unequal resource allocation among the users. The proposed method shows smaller system capacity at lower SNR values because the resources are taken from the users with better SNR values and allotted to the users with worse SNR. From the medium towards the higher SNR values, the proposed method is proven to be better for both resource allocation and overall system capacity. Resource allocation is then analysed between all three users with full usage of the subchannels (FUSC), adaptive modulation and coding (AMC) and the proposed subcarrier permutation in case of different channel quality for every user. The results show the most even resource allocation between the users with the proposed method. The AMC method follows, while the resource allocation is the worst with the FUSC method. When overall system throughput is considered the results show striking domination of the proposed method at lower SNR values and with greater differences in the quality of transmission channels between the users. BER curves also show the domination of the proposed method. Chapter 5 summarizes research results. The proposed method ensures fair resource allocation among users in wider SNR area with varying conditions in the transmission channel compared to WF and PF methods. However, greater fairness in resource allocation causes smaller total system capacity at lower SNR values. The comparative graphs show domination of the proposed method in relation to AMC and FUSC permutations under poor channel conditions. The proposed method enables greater transmission capacity of the system because the resources are allocated to the users based on the best SNR value, adaptive modulation and coding. Additional frequency diversity is achieved owing to the FH. Besides transmission capacity increase, the lower BER values are also achieved for the same reasons. Better characteristics of the proposed method have been, inter alia, achieved due to the needs for greater data quantity transmission. As research results have been promising so far, further research avenues should be targeted towards practical implementation of the proposed method and to signal overhead reduction. -
|b KEY WORDS: adaptive modulation, dynamic resource allocation, frequency hopping, OFDMA, QoS, subcarrier scheduling
|