Traditional digital communication relies on mapping bits to points in a signal constellation, such as Quadrature Amplitude Modulation (QAM) or Phase Shift Keying (PSK). To transmit more data, systems must either increase the density of these constellation grids—requiring higher signal-to-noise ratios (SNR)—or expand the transmission bandwidth.
Index Modulation (IM) sidesteps this paradigm by treating the physical components and structural parameters of a communication system as a combinatorial switchboard. Instead of relying solely on the amplitude or phase of a continuous wave, IM embeds digital information into the activation states or selection indices of available transmission resources. By choosing which antenna, which subcarrier, or which code is active during a given time slot, the system transmits additional data bits without consuming extra power or expanding the spectral footprint.
The Spatial Switchboard
The most intuitive realization of this concept occurs in the spatial domain. In a conventional Multiple-Input Multiple-Output (MIMO) system, multiple radio frequency (RF) chains simultaneously drive an array of antennas to boost data rates through spatial multiplexing. However, this requires expensive, power-hungry hardware for every single antenna element.
Spatial Modulation reimagines this setup. As detailed by Ali Bereyhi et al. in A Fair Comparison Between Spatial Modulation and Antenna Selection in Massive MIMO Systems, the incoming bitstream is split into two parts. One part modulates a standard constellation symbol, while the other part directly selects a specific transmit antenna from the array.
Incoming Bits: [ 1 0 1 1 ]
/ \
[1 0] selects [1 1] maps to a
Antenna Index 2 QPSK Symbol (e.g., +1)
| |
v v
[Antenna 1] ------------> Inactive (0)
[Antenna 2] ------------> Transmits (+1)
[Antenna 3] ------------> Inactive (0)
[Antenna 4] ------------> Inactive (0)
Because the wireless channel acts as a unique spatial signature for each antenna-to-receiver path, the receiver can determine which antenna sent the signal simply by performing channel estimation. This approach offers a profound hardware advantage: as noted by Alice Faisal et al. in Ultramassive MIMO Systems at Terahertz Bands: Prospects and Challenges, spatial modulation minimizes the number of simultaneously active RF components, drastically lowering power consumption in demanding environments like Terahertz-band communication.
Historically, this mapping required the number of transmitters to be a power of two to match binary bit-lengths perfectly. To break this constraint, Jin-Yuan Wang et al. developed a grouping and adaptive bit-mapping scheme in Adaptive Spatial Modulation for Visible Light Communications with an Arbitrary Number of Transmitters. By dynamically partitioning an arbitrary number of transmitters and adjusting the constellation order inversely to the spatial bits, they proved that index modulation can maintain constant data rates across flexible hardware configurations.
Frequency and Cardinality Modulation
The same indexing logic applies to the frequency domain, particularly within Orthogonal Frequency-Division Multiplexing (OFDM). In standard OFDM, every subcarrier carries a data symbol. In OFDM with Index Modulation (OFDM-IM), subcarriers are grouped, and only a subset is activated; the specific pattern of active subcarriers conveys data.
An elegant variation of this technique moves beyond the pattern of activation to the count of active elements. In Enhanced Orthogonal Frequency-Division Multiplexing with Subcarrier Number Modulation, Shuping Dang et al. demonstrate Subcarrier Number Modulation (OFDM-SNM). Instead of mapping bits to specific index permutations, the incoming bit sequence determines the total cardinality of the active subcarrier set.
This cardinality-based approach simplifies the decoding process and enhances reliability. By using instantaneous channel state information to activate only the strongest subcarriers within a block, the authors achieve significant coding gains without sacrificing spectral efficiency.
Taking this concept into multi-user domains, Ziyi Xie et al. demonstrate in Is the Envelope Beneficial to Non-Orthogonal Multiple Access? that integrating index modulation into a power-domain non-orthogonal multiple access (NOMA) framework creates a highly efficient ternary signaling state. Subcarriers are no longer just binary (on/off); they can be inactive, occupied by a single user, or shared by multiple users, leveraging the signal envelope to carry additional data bits.
Generalizing the Index: Codes, Chirps, and Surfaces
The core mechanism of index modulation—conveying information via the selection of a structural parameter—is highly generalizable across different physical domains:
- The Code Domain: In Design of an MISO-SWIPT-Aided Code-Index Modulated Multi-Carrier M-DCSK System for e-Health IoT, Guofa Cai et al. utilize the indices of orthogonal codes to represent data bits, increasing spectral efficiency without expanding signal bandwidth. For multi-user downlinks, Jun Zhao developed a Grouped Code Index Modulation (GrCIM) scheme in 6G Downlink Transmission via Rate Splitting Space Division Multiple Access Based on Grouped Code Index Modulation, mapping private data to the selection of orthogonal Walsh codes to eliminate inter-user interference without consuming extra transmit power.
- The Chirp Domain: For high-mobility communications, Guangyao Liu et al. introduce pre-chirp-domain index modulation in Pre-Chirp-Domain Index Modulation for Affine Frequency Division Multiplexing, mapping data bits directly to the selection of specific pre-chirp parameters within an affine frequency division multiplexing framework.
- The Environment Domain: Index modulation can even turn the physical environment into an active participant. As explored by Ertugrul Basar et al. in Reconfigurable Intelligent Surfaces for 6G: Emerging Hardware Architectures, Applications, and Open Challenges, Reconfigurable Intelligent Surfaces (RIS) can act as secondary transmitters. By dynamically toggling the states of passive reflecting elements to create unique reflection patterns, the surface embeds data into the environment's spatial configuration. This concept is expanded by Shuaishuai Guo et al. in Differential Reflecting Modulation for Reconfigurable Intelligent Surface Based Communications, where information is encoded into the temporal permutation order of pre-defined reflection patterns, bypassing the need for complex channel estimation entirely.
Sparsity as an Information Asset
Why does index modulation work so well in next-generation networks? The answer lies in the mathematical properties of sparse vectors and compressed sensing, particularly when dealing with machine-centric, short-packet communications typical of the Internet of Things (IoT).
In traditional communications, sending a tiny payload requires significant overhead (headers, pilot signals, and channel estimation packets). When the payload is small, this overhead dominates the transmission. Index modulation flips this problem by transforming sparsity into an asset.
As argued by Wonjun Kim et al. in Sparse Vector Transmission: An Idea Whose Time Has Come, encoding information into the positions of non-zero entries of a sparse vector is fundamentally more efficient than traditional amplitude or phase modulation for small payloads.
Traditional: [Header] + [Pilot] + [Data Symbols (Amplitude/Phase)] -> High Overhead
Index-Based: [ 0 0 X 0 0 0 Y 0 ]
^ ^
Position 3 is active Position 7 is active
(The positions *are* the data payload)
This principle is crucial for grant-free massive connectivity. Liang Liu et al. demonstrate in Sparse Signal Processing for Grant-Free Massive Connectivity: A Future Paradigm for Random Access Protocols in the Internet of Things that mapping specific bit strings to unique pilot sequences allows the identity of the detected sequence itself to serve as the data payload. This eliminates the separate data transmission phase entirely, allowing hundreds of IoT devices to be detected and decoded simultaneously using low-complexity sparse reconstruction algorithms.
Key Foundational Work
Cellular architecture and key technologies for 5G wireless communication networks (2014) — contributed 2 concept elements to this direction.
Adaptive Spatial Modulation for Visible Light Communications with an Arbitrary Number of Transmitters (2018) — contributed 2 concept elements to this direction.
Generalized Beamspace Modulation Using Multiplexing: A Breakthrough in mmWave MIMO (2018) — contributed 1 concept elements to this direction.