Advanced Modulation Scheme for Visible Light Communication

In this paper, an improved version of existing modulation schemes is investigated by inserting Hadamard codes (HC) at the peak slot of each symbol for the Visible Light Communication (VLC) system. The performance of the proposed scheme is analyzed and compared with existing schemes On-Off Keying (OOK), Pulse Position Modulation (PPM), Differential PPM (DPPM), Pulse Amplitude Modulation (PAM), Differential Amplitude PPM (DAPPM) and Differential Pulse Interval Modulation (DPIM). Moreover, the symbol error rate (SER) for kth PAM is derived for ds distances between symbols. The performance in Symbol Error Rate (SER) is improved by introducing the Positive and Negative Amplitudes (PNA) levels with respect to DC-bias. The theoretical and simulation results show that the proposed scheme outperforms the existing schemes in terms of bandwidth, and transmission capacity.

techniques are introduced to increase the bit-rate by using Expurgated Pulse-Position Modulation (EPPM) schemes in 12 . The Grouped modulation is proposed in 13 , the multiple-level PAM optical signals are utilized and enhance the data rate up to 50 Mb/s at 1e-6-bit error rate. The trade-off between dimming loss and transmission rate is achieved in 14 by using MPPM. The dimming loss is modeled by the quantization loss within the dimming range for the discrete transmission power levels however, the transmission rate is evaluated by the channel capacity. The proposed technique can be applied to all existing techniques mentioned in the literature.
In this paper, the properties of PPM, DPPM, DAPPM, DPIM and PAM have been analyzed with Hadamard Code (HC), and every HC has a different bandwidth which can be measured from the shortest time of peak slot as given in 15 . The analysis shows that the proposed scheme provides a wider view on the performance for wide range of design parameters. In order to improve SER, the proposed technique does not require an additional power to increase the distance between symbols on a constellation diagram. Therefore, the PNA levels are utilized to maintain average energy avg E . The half numbers of symbols are above and below from P dc on the constellation diagram, and having equal distances on the constellation, as shown in Figure 1. Similarly, the higher values of k can be separated in the same way. However, the larger distances between symbols can cause the flickering problems. Therefore, a tradeoff between SER and flickering is adopted in the room 16 . This paper is organized as follows. Section 2 explains the proposed system model. The analytical analyses are presented in Section 3. Section 4 express the simulation results. Section 5 concludes this paper.

Pdc -a
y t ( ) is the photo-detector current, R represents the photo responsivity of the photo-detector (in A/W). The time average transmitted optical power is given by: The average received optical power generally can be determined as 17 : The Line Of Sight (LOS) links and Non-LOS (NLOS) paths are considered. The DC gain on the first reflection is where, d is the distance between the receiver and transmitter, D 1 is the distance between transmitter and reflective point D 2 is the distance between reflective point and receiver, γ is the reflectance factor, dA is reflective area of small region, α is the angle of irradiance to the receiver, β is the angle of incidence to the receiver, l is the order of Lambertian emission, A PD is the receiving area of photo detector, ϕ is the irradiance angle, ψ is the angle of incidence, The Figure 2 shows that the bit streams are encoded with HC s and Pulse Modulation Schemes (PMS). Then, symbols pass through transmit filter and DC bias for transmission. After passing the AWGN channel, the signal is received at DC filter. Then, it is followed by the set of matched filters to detect the HCs and passes to Maximum Likelihood (ML) for amplitude 1 2 3 , , … ( ) k detection. Finally, the signal is decoded.   Table 1 shows the structure of symbol for two HC codes different schemes. In this paper, the perfect synchronization is considered between the transmitter and receiver and there is no Inter-Symbol Interference (ISI). Vol 12(17) | May 2019 | www.indjst.org

Pulse Position Modulation with Hadamard Code (PPMHC)
Consider a case, when the number of possible symbols can be represented as S L HC PPMHC s = × in PPMHC; where S PPMHC shows the maximum number of symbols, L is the number of slots, and HC s is the number HC. This implies that the bits per symbol M of the scheme can be improved with L and HC, which can be defined as In other word, the number of L is reduced by a factor of 1 /

Differential PPMHC
The differential PPMHC (DPPMHC) is constructed by removing the empty slots of PPM and putting HCs in peak slots of the symbols. Therefore, the average number of slots per symbol can be defined as with the HC as given in Table 1, thus displaying an inherent symbol synchronization capability at the receiver.
Due to variation in the number of slots of a symbol, the error not only affects the corresponding symbol but also subsequent symbols, thus resulting in multiple symbol errors. The AWGN channel, and any given slots L, DPPMHC has the same power requirements per symbol but a slightly higher power requirement in time scale because DPPMHC contains more number of bits at time scale as compared to DPPM, PPM or PPMHC.
Unlike DPPMHC, DPIMHC contains one additional empty slot guard band (GB). Therefore, one or more than one GBs are used depending upon channel condition. The input bit stream is encoded by inserting empty slots in DPPMHC symbol and the bit per symbol is The DPIMHC has a reduced complexity level as compared to PPMHC and DPPMHC due to its built-in symbol synchronization.  Table 1 shows the three bits per symbol by using two pulses HC 2 with one GB. The symbol duration and overall data rate are variable. Therefore, the slot duration is chosen such that the mean symbol duration is equal to the time taken to transmit the same number of bits using fixed symbol length schemes such as OOK or PPMHC. This slot duration is given as.

Differential Amplitude PPMHC (DAPPMHC)
DAPPMHC is the combination of the differential PPM, PAM and HC. It is used to improve the bandwidth efficiency and TC, however, the power of transmission increased with k number of amplitude.

Theoretical Error Performance in Gaussian Channels
The error probability of PPMHC and DPPMHC can be derived for three constellation points which are above and below the DC bias and zero for empty slots on DC bias as shown in Figure 1. The relation between Bit Error Rate (BER) and slot error can be defined as BER SER M L where, E s is the energy of symbols, which encodes log L 2 bits of data.
The probability of slot error for the hard decoding can be derived as: The error probabilities for DPPMHC express as In case of PAM, the PNA levels are transmitted. Therefore, the classic trade-off between SER and flickering is considered as per VLC flickering standards 16 . The general form for k-PAM for interior and end points symbols of constellation diagram can be derived as follows The end points of every high order PAM on constellation is only two but the interior points could be more for the higher order of PAM. The high order PAM is given in Equation It is analyzed in Equation (15) that, the value of d s can be optimized for different modulation schemes to achieve the minimum acceptable flickering in the room. However, the flickering increases as d s , k increase. Moreover, d s is proportion to SER performance and it also producing flickering in the room.

Bandwidth Requirements
The bandwidth is depending upon the number of slots L in specified symbol duration, and number of HCs. An HC Matrix (HCM) 19 The average bandwidth of PPMHC, DPPMHC, DAPPMHC, and DPIMHC are shown in Equation (18), Equation (19), Equation (20) and Equation (21) respectively.

Power Efficiency
The average optical power emitted by an optical wireless LED is limited due to the eye, skin safety and to keep the portable battery powered consumption to a minimum. The performance criteria for each modulation schemes are power and bandwidth efficiency for indoor optical wireless communication systems. Thus, average optical power and bandwidth are required to achieve the desired SER performance or SNR. Mathematically, power efficiency η p is defined as.
where, E pulse , is the energy per pulse and E b is the average energy per bit. The average power requirement for OOK is 20 Therefore, the average power requirement for PPM (soft decision) is approximately 21 The hard decision decoding incurs a 1.5 dB optical power penalty compared the soft decision decoding. The soft decision can only apply to on the fixed length of symbols, for example, PPM and PPMHC. Therefore, only soft decision decoding is considered. The power efficiency of traditional PPM is given in Equation (25).

Simulation Result and Discussion
The setup of the simulation program for indoor VLC system is configured as shown in Figure 2, where the encoder and decoder are used for different modulation schemes. Figure 3 shows the comparison of PPM, PPMHC, and OOK with HCs (OOKHC) in term of average bandwidth requirement for high order modulation. However, the HOM of OOK is achieved by using HC s . Note that, the bandwidth efficiency of PPMHC is better than OOKHC and PPM. Figure 4 is the comparison of bandwidth requirement for DPPM and DPPMHC. The results show that the DPPMHC scheme outperforms the existing scheme DPPM. Note that the average bandwidth requirements for 2 bits per symbol between DPPrMHC 2 and DPPMHC 4 are same because B slot avg ( ) of DPPMHC 4 is greater than DPPMHC 2 . Similarly, in case of DAPPMHC, the average bandwidth of DAPPMHC 2 and DAPPMHC 4 are same at 3 bits/symbols as shown in Figure 5. However, the average bandwidth requirements for DAPPMHC 2 2 is less than DAPPMHC 4 2 at 1 and 2 bits per symbol as shown in Figure 5. Similarly, the average bandwidth requirements for DPPMHC 2 2 is less than DPPMHC 4 2 at 1 bits per symbol as shown in       Figure 9 shows the TC of PPM, DPPM, PPMHC, and DPPMHC at HOM. As depicted in Figure 9, the TC of DPPM, PPMHC 2 and DPPMHC 2 , PPMHC 4 are similar at M ≥ 8. However, the TC of DPPMHC 4 is the double of DPPMHC 2 , and PPMHC 4 and, TC of DPPMHC 2 and PPMHC 4 is double of DPPM, PPMHC 2 at M ≥ 8 . The TC can be further improving by including PAM as shown in Figure 10. Figure 11 is showing the SER performance at different values of d s , however, the average transmitted power is same because PNA levels technique is used. The higher values of d s have better SER performances but the flickering problem may occur in the room.

Conclusion
In this paper, the improved modulation schemes are presented HCs are combined with PPM, DPPM, and DAPPM to improve the transmission rates, spectral efficiency, and power efficiency. This allows the transmission of more information and better performance at a certain number of orthogonal pulses when compared with conventional modulation techniques. The bandwidth, power, and TC are improved by combining with existing modulation schemes. The HCM generate multiple orthogonal pulses at the cost of higher bandwidth but in this paper, the results show that the proposed scheme outperforms the existing schemes in term of bandwidth, TC, and power efficiency.