1 | Page Advance in Adaptive Modulation for Fading Channels. Jehad M.A. Hamamreh, Hassan El_sallabi, Mohamed Abdallah, Khalid Qaraqe. Department of Electrical and computer Engineering, Texas A&M Univ. at Qatar (TAMUQ) {jehad.hamamra, hassan.el_sallabi, mohamed.abdallah, khalid.qaraqe }@ qatar.tamu.edu Abstract — This work proposes an advance in advanced modulation technique (AMT) that is based on utilizing more channel state information in addition to signal to noise ration (SNR) ranges. The particular new information is related to how sever fading the channel experiences. The amount of severity in this work is measured with amount of fading (AF). This additional information helps to distinguish between different channel conditions that have same average set of SNR ranges but different levels of fading severity that may be utilized to increase performance of conventional adaptive modulation technique (CAMT). The different levels of fading severity and similar sets of SNR ranges have been tested with Nakagami-m fading channels. The AF measure of fading severity is equal to 1/m in this radio fading channels. So, the investigation in this work is based on testing how to leverage the AF dimension in addition to the conventional approach used in CAMT. In this work we show that the bit error rate (BER) of different modulation schemes with M order depends on fading amount for every range of SNR defined by sets of AMT. The switching method that we build is based on our new derived formula, that combines SNR, BER, M and m=1/AF with each other in one relationship. Results show not only exact prediction for which modulation order (M) should be used at each AF, SNR, and BER values but rather dramatic improvements in BER performance and throughput when AF is leveraged with SNRs range set approach defined in CAMT. Utilization of AF with SNR ranges allows adapting higher modulation order in channel conditions that were not possible with conventional AMT. Index Terms— Modulation order, Nakagami-m Fading Channel, CAMT, AMT, SNR, BER, throughputs. (key words) I. INTRODUCTION Smart phones are becoming dominant handsets that are available to wireless technology users. Wireless access to internet is default scenario with vast majority of internet users. The increasing demand for high speed wireless internet services makes current technologies meet their limit due to channel impairments. The conventional adaptive modulation technique (CAMT) is no longer helpful due to high data rate requirements of new technologies and wireless video streaming in addition to other applications such as downloading large files. The CAMT is one of the developed approaches that are considered powerful techniques that are currently being used in advanced wireless communication systems such as long term evolution (LTE) technology. This technique is used to enhance energy efficiency and increase spectral efficiency of wireless communication systems over fading channels. The CAMT dynamically changes modulation schemes based on channel conditions to maximize throughput with minimum bit error rate (BER) based on channel state information of each user, which is sent back to transmitter by receiver side via reliable channel. The CAMT is based on predefined set of ranges signal to noise ratios (SNRs) for different orders of modulation schemes. The more increase in SNRs lead to higher level of ranges of SNRs set that lead to a higher modulation order. This will allow to higher transmission speed utilizing the good channel condition. In order to minimize BER, when channel condition degrades, modulation order is reduced, which result to lower spectral efficiency but more robust modulation scheme. The dynamicity of changing modulation order based on SNRs ranges of radio channel is the key part of CAMT to increase throughput and minimize the BER. However, the performance of communication systems, in terms of throughput, average symbol error rate, is strongly influenced by channel model characteristics. There is variety of versatile channel characteristics in our contemporary wireless systems. The most commonly utilized channel characteristic is the multipath fading proving to be the main hindrance for wireless mobile radio to achieve reliable communication [1-2]. Previous studies have shown that one of the most widely accepted statistical model that covers a wide range of fading scenarios is Nakagami-m distribution [2]. Different kinds of fading conditions may be created just by varying fading index (m) of Nakagami-m channel. Some of these fading scenarios include Rayleigh (m=1), one-sided Gaussian (m=0.5), severe fading (0.5<m<1) and small fading (m>10). The performance of different communication systems with various modulation types using m-Nakagmi channel are investigated and studied [3-8]. We have to take care of these fading indexes while designing a complete wireless system. Traditional early researches on adaptive modulation topic [9-14] show that it is essential in order to produce an efficient wireless system design to have optimized system performance curves (BER, throughputs) for different modulation techniques, that enable us to switch between different modulation based on the SNR level. However, we have presented here that not only SNR affects on switching method of adaptive modulation, but also the fading index (m), we derived a new equation combining the relationship between SNR, BER, M and m. Based on the obtained results from curves and equations, a new selection method for adaptive modulation is introduced which is based on the amount of fading along with SNR leading to dramatic