This is a hard question to answer because the terms mean different things to different people.

Originally, the term LIS was used to describe large active antenna arrays, like massive MIMO but much larger: Article Beyond Massive-MIMO: The Potential of Data-Transmission with...

Hence, a LIS generates, transmits, and receives signals. The arrays are not only large but also consisting of very small elements, giving an almost continuous control over the electromagnetic waves over the surface.

In contrast, a RIS is receiving a signal from another location and scatters it in a controllable way. It is like a (semi-)passive large array.

Unfortunately, some researchers have used LIS as more-or-less synonymous with RIS, which causes confusion.

Thanks a lot, Professor Emil Björnson

Indeed, it is a difficult question. I realized this question after I read a brief explanation of their difference in a paper Article Asymptotic Max-Min SINR Analysis of Reconfigurable Intellige...

It seems people usually do not distinguish them, in terms of mathematical modeling. So, do you think I can have same anticipation for their phase shift?

If I want them to focus a beam to a constant location, I conclude that using RIS would be a better choice, as it seems it uses less power. Is this right?

Can I model their power consumption in a different way?

I really appreciate your help.

Channel modeling is dictated by the specific system realization with a focus on the application and is further dictated by the physical constraints set by the environment, the device, and the users. Then, as soon you have a proper definition of what is meant by Large Intelligent Surfaces (LIS) and Reconfigurable Intelligent Surfaces (RIS) and their modus operandi, you will then be able to work out what characteristics/properties are the most relevant. I think this is a rather general principle for channel modeling.

Sir Emil Björnson will you please explain or refer to your any article for simple channel modelling of IRS? If we have signal sent from BS (mimo) to IRS and IRS to User. How we model this?

Haleema Sadia Overwhelming number of papers usually focus on single antenna devices owing to the analytical tractability. Also, since the research in this domain is still considered to be in its nascent stage, it is still evolving to unearth new findings. Nevertheless, you can refer, for example, to a paper titled "Spectral and Energy Efficiency of IRS-Assisted MISO Communication with Hardware Impairments" for what you sought.

Sourabh Solanki Thank you for your answer and reference. I have already gone through this article.

Dear Masoud Arash ,

The main objectives to add such devices as LIS and RIS is to configure and control the wireless communication channels between the source and the destination. Up till now, one treats the channel as a completely physical identity that one can describe it and model it but not configure and control it.

With using LIS and RIS the wireless communication cannel characteristics can be reformed. The basic idea behind such surfaces is to control constructive and destructive interferes at the location of the receiver. Such surfaces are able to to divide incident radiations on them and reflect and phase shift them to maximize the signal at the intended receiver while minimize the interference to the unintended receiver.

Any signal path is always modeled by a transmission coefficient and specific phase shift.

I would like that you refer to this link:https://www.comsoc.org/publications/best-readings/reconfigurable-intelligent-surfaces

Best wishes